The two basic objectives of this project are to: 1) determine which of the pharmacologic agents currently being tested in vitro and in vivo in laboratory models of lung injury are effective in patients with acute lung injury or who are predisposed to develop lung injury and 2) to establish the utility of the multiple indicator dilution technique for measurement of the permeability surface area product (PS) of the lung in patients with both clinical and subclinical acute lung injury as a measure of severity of lung injury and a tool to assess probable outcome. These objectives are being pursued in tandem because the clinical markers for the detection of acute lung injury and its resolution are imprecise and do not yield information which is prognostically useful. In patients undergoing hemodialysis (a controlled model of "lung injury") we will measure PS and capillary volume by gas inhalation techniques in an effort to determine the contribution of surface area to PS and to derive better estimates of permeability. PS measurements along with the routine physiologic measurements will be used in clinical trails of pharmacologic agents in ARDS and in clinical circumstances known to predispose to ARDS (e.g. sepsis). Effects of pulmonary vasodilation on PS, gas exchange and hemodynamics will be studied in similar patients in order to determine the role of reversible vascular derecruitment in the pathophysiology of ARDS. The efficacy of the nonsteroidal anti- inflammatory agent, ibuprofen, will be studied in septic patients initially then in patients with ARDS if found safe. The effects of N-acetylcysteine (a free radical scavenger) will be tested in pilot studies in patients with established ARDS. Other agents which may be given consideration for clinical trials include PGE2, aerosolized catalase and superoxide dismutase, and allopurinol. These studies promise new insights into the pathophysiology of acute lung injury in humans, new tools for defining abnormalities in lung function and predicting outcome and identification of effective pharmacologic interventions in the treatment of ARDS.